Flask cells of an inverting wild-type Volvox embryo (left) and an inversionless mutant (right). Arrowheads indicate cytoplasmic bridges that link neighbouring cells. In the wild-type embryo, the cells have moved so that they are linked by bridges at their narrow ends. However, the mutant cells have failed to move, so they remain linked near their wide ends. Figure adapted from Nishii et al. © (2003) with permission from Elsevier Science.

Curling of a sheet of cells is a common morphogenetic process used at several developmental stages, including gastrulation and neurulation. A simplified, analogous process occurs during development of Volvox carteri, a multicellular green alga in which a spherical monolayer of cells must turn itself inside out. Recent work by Kirk and colleagues (Cell 113, 743–753 (2003)) has identified a new kinesin microtubule motor that is necessary for generating the cell movements required for this inversion process.

At the time of inversion, the spherical monolayer of cells is interconnected by a band of cytoplasmic bridges. The monolayer is made of flask-shaped cells with their wide ends facing into the interior of the sphere and their narrow ends facing outwards. The plane of the cytoplasmic bridge network connects these cells near their wide ends. Curling is initiated by cells moving with respect to their cytoplasmic bridge such that they go from being connected at their wide ends to being connected at their narrow ends (see arrows in the left image). This movement forces the cell sheet to curl sharply outwards and back over itself.

Mutants in the inversion process were identified 30 years ago, but because inversion is necessary to acquire the adult configuration (that is, reproductive cells on the interior) inversionless mutants are unable to mate, and so normal Mendelian analysis could not be used to identify the genes responsible. Now, Kirk and colleagues use the Jordan temperature-sensitive transposon to generate transposon-tagged mutants, which can then be used to clone the genes involved. One of the inversion mutants identified, InvA, was found to have a defect in cell movements such that the cytoplasmic bridges remain at the wide end of the cells (see arrowheads in the image on the right), and, thus, the curling process does not occur. By sequence analogy, the identified gene was predicted to be a plus-end-directed kinesin microtubule motor. InvA is expressed at the time of inversion and the protein is localized to the cytoplasmic bridge.

Identification of a microtubule motor gave immediate insight into the mechanism of inversion, because a network of microtubules running the length of the flask-shaped cells and passing just adjacent to the cytoplasmic bridge were known to be important for the inversion process.

From these observations, the authors propose a model for how InvA might generate the necessary cell movements. Microtubules are aligned with their plus ends terminating at the narrow end of the cell. Thus, the plus end motor activity of InvA would move it towards the narrow end of the cell. However, if InvA is attached to the cytoplasmic bridge structure, as suggested by its localization, this motor movement would actually tend to move the microtubule network in the opposite direction. This could function to move the cell contents toward the wide end of the cell, leaving the cytoplasmic bridge connecting cells at their narrow tips — the correct confirmation to initiate inversion.

More work will be required to determine if this is indeed the mechanism. It will also be interesting to determine if analogous mechanisms are used in gastrulation or neurulation. The simplicity of Volvox inversion, combined with the new genetic tools for identifying mutant genes, should be fertile ground for identifying more of the molecular players in this prototypical morphogenetic event.